Artificial Intelligence and Coronavirus

Author Name: Maheen Shafeeq      19 Apr 2020     Emerging Technologies

Coronavirus (Covid-19) has already had an immense impact on the world by infecting over two million persons and causing more than one hundred thousand deaths in just a few months. Given the suffering unleashed by this pandemic, researchers and innovators are mobilizing technological resources towards rescuing humanity from the global health crisis.

Although the world has been exposed to numerous pandemics in the past, the technological resources available to grapple with those outbreaks were comparatively limited. In recent years, by contrast, advancements in Artificial Intelligence (AI) have begun to show significant promise in the public health domain.

Prompted by the urgency to eradicate coronavirus, technology giants are using this valuable resource to assist the healthcare sector by helping medical researchers scour through tens of thousands of relevant research papers.

Since the virus was first reported in December 2019, over 2000 papers have been published on the subject. As it is difficult for any group of humans to manually peruse all these documents within a short timeframe, AI algorithms have been trained to funnel the appropriate information.

As an example, Microsoft research, the National Library of Medicine and the Allen Institute for AI (AI2) had gathered over 29,000 papers related to the wider coronavirus family; from which the AI data analysis systems processed 13,000 to draw useful information on the virus.

In that instance, all the information was then used to answer ten key questions about the virus, ranging from understanding the genetic and physical properties of the virus (for vaccine development), to risk analysis and treatment without drugs.

AI is also playing an important role in bioinformatics to help medical researchers understand the structure of the virus to develop a vaccine. The delay in the vaccine for the coronavirus, they say, is because it is challenging to understand the structure of coronavirus as it belongs to a family of enveloped viruses that contain single-strand Ribonucleic acid (RNA). Due to the single strand, the virus can mutate rapidly. The mutation alters the structure of RNA, which then makes it difficult to analyze the virus to develop the vaccine.

As coronavirus is spreading all over the world, the rapid rate of mutation is making it stronger and more complicated to understand through time-consuming experiments. Therefore, to speed up the research, AI arm of technology giants such as DeepMind of Google and LinearFold of Baidu, the largest Chinese search engine, created faster algorithms to understand this complex virus.

These algorithms only use a fraction of time and memory compared to existing methodologies to develop vaccines. As found out by Baidu the AI method reduces the overall analysis time from 55 minutes to 27 seconds, which makes the AI process 127 times faster than the traditional method. These algorithms use different modelling methods to help predict the structure of the virus. Once the structure of the virus is understood by medical researchers, they will be able to draft the genetic makeup of Covid-19, which will make vaccine discovery, somewhat quicker and simpler.

Despite these efforts, it is important to note that the discovery of the vaccine for coronavirus is still only at its initial stage. A lack of funding, understanding and resources is also a major hurdle in the wider scheme of things. Donald Trump’s recent threat to withhold funding for thw World Health Organization (WHO) is indicative of political and governance issues that also loom large.

Nevertheless, AI can play a crucial role in the surveillance of the virus. The example of how China rolled out artificial intelligence to respond to the virus, will serve as a guiding toolkit for the rest of the world.

Firstly, the Chinese government installed AI-based thermal scanners to monitor potential patients in places such as grocery stores, medical dispensaries, and other functional public zones. This can be help identify the potential patients that are not practicing self-quarantine and should be taken under hospital care. Secondly, the Chinese government used its most advanced facial recognition technology to keep track of the affected people and those who recovered.

Lastly, AI has enabled the Chinese doctors to make an informed decision on who has a better chance of survival or the highest risk of death among the competing coronavirus patients. They have done this by analyzing the blood samples of coronavirus patients since the date of their admission. Taking blood samples at the appropriate time will also help the rest of the world’s doctors to predict necessary measures.

While the progress towards the discovery of a cure is still early, AI and technological advancement have been extremely valuable to monitor and control the virus; and have shown great potential to have a meaningful impact from the pandemic.

 

Maheen Shafeeq is a research fellow at Centre for Aerospace and Security Studies. This article was first published at regionaltimes http://regionaltimes.com/19Apr2020/4.jpg. She can be reached at cass.thinker@gmail.com

 

Recent Articles